U.S. patent application number 12/397505 was filed with the patent office on 2009-09-10 for method and apparatus for encoding/decoding multi-channel audio signal by using a plurality of variable length code tables.
This patent application is currently assigned to SAMSUNG ELECTRONICS CO., LTD.. Invention is credited to Jong-hoon JEONG, Chul-woo LEE, Geon-hyoung LEE, Nam-suk LEE, Han-gil MOON.
Application Number | 20090228284 12/397505 |
Document ID | / |
Family ID | 41054559 |
Filed Date | 2009-09-10 |
United States Patent
Application |
20090228284 |
Kind Code |
A1 |
MOON; Han-gil ; et
al. |
September 10, 2009 |
METHOD AND APPARATUS FOR ENCODING/DECODING MULTI-CHANNEL AUDIO
SIGNAL BY USING A PLURALITY OF VARIABLE LENGTH CODE TABLES
Abstract
A method and apparatus for encoding a multi-channel audio signal
is provided. The method includes the operations of selecting a
first variable length code table (VLC) that is to be used to encode
the multi-channel audio signal in a given frequency band, from
among a plurality of VLC tables, and encoding the multi-channel
audio signal by using the first VLC table. Thus, the multi-channel
audio signal can be encoded by using VLC tables that differ
according to frequency bands.
Inventors: |
MOON; Han-gil; (Seoul,
KR) ; LEE; Geon-hyoung; (Hwaseong-si, KR) ;
LEE; Chul-woo; (Suwon-si, KR) ; JEONG; Jong-hoon;
(Suwon-si, KR) ; LEE; Nam-suk; (Suwon-si,
KR) |
Correspondence
Address: |
SUGHRUE MION, PLLC
2100 PENNSYLVANIA AVENUE, N.W., SUITE 800
WASHINGTON
DC
20037
US
|
Assignee: |
SAMSUNG ELECTRONICS CO.,
LTD.
Suwon-si
KR
|
Family ID: |
41054559 |
Appl. No.: |
12/397505 |
Filed: |
March 4, 2009 |
Current U.S.
Class: |
704/500 |
Current CPC
Class: |
G10L 19/035
20130101 |
Class at
Publication: |
704/500 |
International
Class: |
G10L 19/00 20060101
G10L019/00 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 4, 2008 |
KR |
10-2008-0020068 |
Claims
1. A method of encoding a multi-channel audio signal, the method
comprising: selecting a first variable length code (VLC) table that
is to be used to encode the multi-channel audio signal in a first
frequency band, from among a plurality of VLC tables; and encoding
the multi-channel audio signal by using the first VLC table,
wherein the plurality of VLC tables correspond to a plurality of
frequency bands including the first frequency band, and wherein, if
the multi-channel audio signal is in a second frequency band, a
second VLC table among the plurality of VLC tables is used to
encode the multi-channel audio signal.
2. The method of claim 1, wherein the encoding of the multi-channel
audio signal comprises encoding at least one of an intensity and a
parameter of the multi-channel audio signal in the first frequency
band by using the first variable length code table.
3. The method of claim 1, wherein the variable length code tables
are Huffman code tables.
4. The method of claim 3, wherein the multi-channel audio signal is
a stereo-audio signal.
5. The method of claim 4, wherein the parameter of the
multi-channel audio signal is a parameter for determining an
intensity of a left-channel audio signal in the first frequency
band and an intensity of a right-channel audio signal in the first
frequency band.
6. The method of claim 5, wherein the parameter for determining the
intensities of the left-channel audio signal and the right-channel
audio signal is information about one of an angle between a third
vector associated with an intensity of a mono-audio signal and a
first vector associated with an intensity of the left-channel audio
signal, or an angle between the third vector and a second vector
associated with an intensity of the right-channel audio signal in a
vector space formed so that the first and second vectors form a
given angle.
7. The method of claim 6, wherein the information about the angle
is one of a cosine value of the angle between the first and third
vectors and a cosine value of the angle between the second and
third vectors.
8. The method of claim 1, wherein the frequency band is a
sub-band.
9. A method of decoding a multi-channel audio signal, the method
comprising: selecting a first variable length decode (VLD) table
that is to be used to decode the multi-channel audio signal in a
fist frequency band, from among a plurality of VLD tables; and
decoding the multi-channel audio signal by using the first VLD
table, wherein the plurality of VLD tables correspond to a
plurality of frequency bands including the first frequency band,
and wherein, if the multi-channel audio signal is in a second
frequency band, a second VLD table among the plurality of VLD
tables is used to decode the multi-channel audio signal.
10. The method of claim 9, wherein the decoding of the
multi-channel audio signal comprises decoding at least one of an
intensity and a parameter of the multi-channel audio signal in the
first frequency band by using the first variable length code
table.
11. The method of claim 9, wherein the variable length code tables
are Huffman code tables.
12. The method of claim 11, wherein the multi-channel audio signal
is a stereo-audio signal.
13. The method of claim 12, wherein the parameter of the
multi-channel audio signal is a parameter for determining an
intensity of a left-channel audio signal in the first frequency
band and an intensity of a right-channel audio signal in the first
frequency band.
14. The method of claim 13, wherein the parameter for determining
the intensities of the left-channel audio signal and the
right-channel audio signal is information about one of an angle
between a third vector associated with an intensity of a mono-audio
signal and a first vector associated with an intensity of the
left-channel audio signal, or an angle between the third vector and
a second vector associated with an intensity of the right-channel
audio signal in a vector space formed so that the first and second
vectors form a given angle.
15. The method of claim 14, wherein the information about the angle
is one of a cosine value of the angle between the first and third
vectors and a cosine value of the angle between the second and
third vectors.
16. The method of claim 9, wherein the frequency band is a
sub-band.
17. An apparatus for encoding a multi-channel audio signal, the
apparatus comprising: a control unit that selects a first variable
length code (VLC) table that is to be used to encode the
multi-channel audio signal in a first frequency band, from among a
plurality of VLC tables; and an encoding unit that encodes the
multi-channel audio signal by using the first VLC table, wherein
the plurality of VLC tables correspond to a plurality of frequency
bands including the first frequency band, and wherein, if the
multi-channel audio signal is in a second frequency band, the
control unit selects a second VLC table among the plurality of VLC
tables to encode the multi-channel audio signal.
18. The apparatus of claim 17, wherein the encoding unit encodes an
intensity or a parameter of the multi-channel audio signal in the
first frequency band by using the first variable length code
table.
19. The apparatus of claim 17, wherein the variable length code
tables are Huffman code tables.
20. The apparatus of claim 19, wherein the multi-channel audio
signal is a stereo-audio signal.
21. The apparatus of claim 20, wherein the parameter of the
multi-channel audio signal is a parameter for determining an
intensity of a left-channel audio signal in the first frequency
band and an intensity of a right-channel audio signal in the first
frequency band.
22. The apparatus of claim 21, wherein the parameter for
determining the intensities of the left-channel audio signal and
the right-channel audio signal is information about one of an angle
between a third vector associated with an intensity of a mono-audio
signal and a first vector associated with an intensity of the
left-channel audio signal, or an angle between the third vector and
a second vector associated with an intensity of the right-channel
audio signal in a vector space formed so that the first and second
vectors form a given angle.
23. The apparatus of claim 22, wherein the information about the
angle is one of a cosine value of the angle between the first and
third vectors and a cosine value of the angle between the second
and third vectors.
24. An apparatus for decoding a multi-channel audio signal, the
apparatus comprising: a control unit that selects a first variable
length decode (VLD) table that is to be used to decode the
multi-channel audio signal in a first frequency band, from among a
plurality of VLD tables; and a decoding unit that decodes the
multi-channel audio signal by using the first VLD table. wherein
the plurality of VLD tables correspond to a plurality of frequency
bands including the first frequency band, and wherein, if the
multi-channel audio signal is in a second frequency band, the
control unit selects a second VLD table among the plurality of VLD
tables to decode the multi-channel audio signal.
25. The apparatus of claim 24, wherein the decoding unit decodes an
intensity or a parameter of the multi-channel audio signal in the
first frequency band by using the first variable length code
table.
26. The apparatus of claim 24, wherein the variable length code
tables are Huffman code tables.
27. The apparatus of claim 26, wherein the multi-channel audio
signal is a stereo-audio signal.
28. The apparatus of claim 27, wherein the parameter of the
multi-channel audio signal is a parameter for determining an
intensity of a left-channel audio signal in the first frequency
band and an intensity of a right-channel audio signal in the first
frequency band.
29. The apparatus of claim 28, wherein the parameter for
determining the intensities of the left-channel audio signal and
the right-channel audio signal is information about one of an angle
between a third vector associated with an intensity of a mono-audio
signal and a first vector associated with an intensity of the
left-channel audio signal, or an angle between the third vector and
a second vector associated with an intensity of the right-channel
audio signal in a vector space formed so that the first and second
vectors form a given angle.
30. The apparatus of claim 29, wherein the information about the
angle is one of a cosine value of the angle between the first and
third vectors and a cosine value of the angle between the second
and third vectors.
31. A computer readable recording medium having recorded thereon a
program for executing the method of claim 1.
32. A computer readable recording medium having recorded thereon a
program for executing the method of claim 9.
33. The method of claim 1, wherein a number of the plurality of the
VLC tables is (n.times.(p+1)).times.m, where n is a number of the
frequency bands that the multi-channel audio signal can be
transmitted in; p is a number of parameters used for encoding a
multi-channel audio signal, and m is a number of channels of the
multi-channel audio signal, wherein n.times.m number of VLC tables
among the plurality of the VLC tables are provided for encoding
intensities of the multi-channel audio signal, and wherein the
parameters comprise information on at least one of intensity
difference, correlation difference and phase difference between the
channels of the multi-channel audio signal.
34. The method of claim 9, wherein a number of the plurality of the
VLD tables is (n.times.(p+1)).times.m, where n is a number of the
frequency bands that the multi-channel audio signal can be
transmitted in; p is a number of parameters used for decoding a
multi-channel audio signal, and m is a number of channels of the
multi-channel audio signal, wherein n.times.m number of VLD tables
among the plurality of the VLD tables are provided for decoding
intensities of the multi-channel audio signal, and wherein the
parameters comprise information on at least one of intensity
difference, correlation difference and phase difference between the
channels of the multi-channel audio signal.
35. The apparatus of claim 17, wherein a number of the plurality of
the VLC tables is (n.times.(p+1)).times.m, where n is a number of
the frequency bands that the multi-channel audio signal can be
transmitted in; p is a number of parameters used for encoding a
multi-channel audio signal, and m is a number of channels of the
multi-channel audio signal, wherein n.times.m number of VLC tables
among the plurality of the VLC tables are provided for encoding
intensities of the multi-channel audio signal, and wherein the
parameters comprise information on at least one of intensity
difference, correlation difference and phase difference between the
channels of the multi-channel audio signal.
36. The apparatus of claim 24, wherein a number of the plurality of
the VLD tables is (n.times.(p+1)).times.m, where n is a number of
the frequency bands that the multi-channel audio signal can be
transmitted in; p is a number of parameters used for decoding a
multi-channel audio signal, and m is a number of channels of the
multi-channel audio signal, wherein n.times.m number of VLD tables
among the plurality of the VLD tables are provided for decoding
intensities of the multi-channel audio signal, and wherein the
parameters comprise information on at least one of intensity
difference, correlation difference and phase difference between the
channels of the multi-channel audio signal.
Description
CROSS-REFERENCE TO RELATED PATENT APPLICATION
[0001] This application claims priority from Korean Patent
Application No. 10-2008-0020068, filed on Mar. 4, 2007, in the
Korean Intellectual Property Office, the disclosure of which is
incorporated herein in its entirety by reference.
BACKGROUND OF THE INVENTION
[0002] 1. Field of the Invention
[0003] Methods and apparatuses consistent with the present
invention relate to encoding/decoding of a multi-channel audio
signal, and more particularly, to encoding/decoding of a
multi-channel audio signal in consideration of probability
distributions of an intensity or parameters of the multi-channel
audio signal which differ according to frequency bands.
[0004] 2. Description of the Related Art
[0005] Examples of a general multi-channel audio encoding method
include waveform audio coding and parametric audio coding. Examples
of waveform audio coding include Moving Picture Experts Group-2
(MPEG-2) multi-channel audio coding, Advanced Audio Coding (AAC)
for multi-channel, Bit Sliced Arithmetic Coding (BSAC)/Audio Video
Coding Standard (AVS) multi-channel audio coding, etc.
[0006] In parametric audio coding, an audio signal is divided into
components, such as frequencies or amplitudes, in a frequency
domain, and information about the frequencies, amplitudes, or the
like is parameterized, thereby encoding the audio signal. For
example, when a stereo audio signal is encoded by parametric audio
coding, a left-channel audio signal and a right-channel audio
signal of the stereo audio signal are down-mixed to generate a
mono-audio signal, and the mono-audio signal is encoded. Then,
parameters of each frequency band, such as, an interchannel
intensity difference (IID), an interchannel correlation (IC), an
overall phase difference (OPD), and an interchannel phase
difference (IDP) are encoded. A plurality of frequency bands are
obtained by dividing a frequency domain into a plurality of areas.
During parametric audio coding, an encoding side encodes the
aforementioned parameters of each frequency band and transmits the
encoded parameters to a decoding side. The decoding side decodes
the encoded parameters in order to restore the original parameters,
and transforms a mono-audio signal into a stereo-audio signal on
the basis of the restored parameters.
[0007] According to a related art, a multi-channel audio signal is
encoded without considering a characteristic of a multi-channel
audio signal which varies according to frequency bands. Thus, the
multi-channel audio signal is inefficiently encoded.
SUMMARY OF THE INVENTION
[0008] The present invention provides a method and apparatus for
encoding/decoding a multi-channel audio signal in consideration of
a characteristic of the multi-channel audio signal, which varies
according to frequency bands, and a computer readable recording
medium having recorded thereon a program for executing the
method.
[0009] According to an aspect of the present invention, there is
provided a method of encoding a multi-channel audio signal in a
given frequency band, the method comprising the operations of
selecting a first variable length code (VLC) table that is to be
used to encode the multi-channel audio signal in the given
frequency band, from among a plurality of VLC tables, and encoding
the multi-channel audio signal by using the first VLC table.
[0010] The operation of encoding the multi-channel audio signal may
comprise encoding at least one of an intensity and a parameter of
the multi-channel audio signal in the given frequency band by using
the first VLC table.
[0011] According to another aspect of the present invention, there
is provided a method of decoding a multi-channel audio signal in a
given frequency band, the method comprising the operations of
selecting a first variable length decode (VLD) table that is to be
used to decode the multi-channel audio signal in the given
frequency band, from among a plurality of VLD tables; and decoding
the multi-channel audio signal by using the first VLD table.
[0012] The operation of decoding the multi-channel audio signal may
comprise decoding at least one of an intensity and a parameter of
the multi-channel audio signal in the given frequency band by using
the first VLD table.
[0013] According to another aspect of the present invention, there
is provided an apparatus for encoding a multi-channel audio signal
in a given frequency band, the apparatus comprising: a control unit
selecting a first VLC table that is to be used to encode the
multi-channel audio signal in the given frequency band, from among
a plurality of VLC tables; and an encoding unit encoding the
multi-channel audio signal by using the first VLC table.
[0014] The encoding unit may encode at least one of an intensity
and a parameter of the multi-channel audio signal in the given
frequency band by using the first VLC table.
[0015] According to another aspect of the present invention, there
is provided an apparatus for decoding a multi-channel audio signal
in a given frequency band, the apparatus comprising: a control unit
selecting a first VLD table that is to be used to decode the
multi-channel audio signal in the given frequency band, from among
a plurality of VLD tables; and a decoding unit decoding the
multi-channel audio signal by using the first VLD table.
[0016] The decoding unit may decode at least one of an intensity
and a parameter of the multi-channel audio signal in the given
frequency band by using the first VLD table.
[0017] The VLC or VLD tables may be Huffman code tables.
[0018] The multi-channel audio signal may be a stereo-audio signal.
The parameter of the multi-channel audio signal may be a parameter
for determining an intensity of a left-channel audio signal in the
given frequency band and an intensity of a right-channel audio
signal in the given frequency band.
[0019] The parameter for determining the intensities of the
left-channel audio signal and the right-channel audio signal may be
information about either an angle between a third vector associated
with an intensity of a mono-audio signal and a first vector
associated with an intensity of the left-channel audio signal or an
angle between the third vector and a second vector associated with
an intensity of the right-channel audio signal in a vector space
formed so that the first and second vectors form a given angle.
[0020] The information about the angle may be either a cosine value
of the angle between the first and third vectors or a cosine value
of the angle between the second and third vectors.
[0021] According to another aspect of the present invention, there
is provided a computer readable recording medium having recorded
thereon a program for executing the above-described method of
encoding/decoding a multi-channel audio signal.
BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The above and other aspects of the present invention will
become more apparent by describing in detail exemplary embodiments
thereof with reference to the attached drawings, in which:
[0023] FIG. 1 illustrates an apparatus for encoding a multi-channel
audio signal, according to an exemplary embodiment of the present
invention;
[0024] FIG. 2 illustrates frequency bands according to an exemplary
embodiment of the present invention;
[0025] FIG. 3 illustrates an apparatus for encoding a multi-channel
audio signal, according to another exemplary embodiment of the
present invention;
[0026] FIG. 4 is a diagram for explaining a method of encoding a
parameter for determining the intensities of a left-channel audio
signal and a right-channel audio signal, according to an exemplary
embodiment of the present invention;
[0027] FIGS. 5A and 5B are graphs illustrating probability
distributions of an angle between vectors associated with
intensities of a right-channel audio signal and a mono-audio
signal, according to an exemplary embodiment of the present
invention;
[0028] FIGS. 6A and 6B illustrate variable length code (VLC) tables
according to an exemplary embodiment of the present invention;
[0029] FIG. 7 is a flowchart illustrating a method of encoding a
multi-channel audio signal, according to an exemplary embodiment of
the present invention;
[0030] FIG. 8 illustrates an apparatus for decoding a multi-channel
audio signal, according to an exemplary embodiment of the present
invention;
[0031] FIG. 9 illustrates an apparatus for decoding a multi-channel
audio signal, according to another exemplary embodiment of the
present invention; and
[0032] FIG. 10 is a flowchart of a method of decoding a
multi-channel audio signal, according to an exemplary embodiment of
the present invention.
DETAILED DESCRIPTION OF EXEMPLARY EMBODIMENTS
[0033] The present invention will now be described more fully with
reference to the accompanying drawings, in which exemplary
embodiments of the invention are shown.
[0034] FIG. 1 illustrates an apparatus 100 for encoding a
multi-channel audio signal, according to an exemplary embodiment of
the present invention. Referring to FIG. 1, the apparatus 100 for
encoding the multi-channel audio signal includes a control unit 110
and an encoding unit 120.
[0035] The control unit 110 selects a variable length code (VLC)
table that is to be used to encode a multi-channel audio signal in
a given frequency band, from among a plurality of VLC tables. In
the exemplary embodiment of the present invention, encoding is
performed using a characteristic of a multi-channel audio signal
which has different values according to frequency bands. The
characteristic of the multi-channel audio signal that varies
according to frequency bands may be a probability distribution of
an intensity value or a parameter value of the multi-channel audio
signal.
[0036] The intensity of a multi-channel audio signal denotes an
intensity of the multi-channel audio signal in a given frequency
band on a generated power spectrum when the multi-channel audio
signal is transformed into the frequency domain by fast Fourier
transformation or the like. The parameter of the multi-channel
audio signal denotes any parameter generated while encoding the
multi-channel audio signal. Examples include an interchannel
intensity difference (IID), an interchannel correlation (IC), an
overall phase difference (OPD), and an interchannel phase
difference (IDP), which have been described above with reference to
a related art.
[0037] When illustrating the intensity of a multi-channel audio
signal, a probability, that the intensity of the multi-channel
audio signal in a low frequency band slightly varies according to
time, is higher than a probability that the intensity of the
multi-channel audio signal in a high frequency band slightly varies
according to time. In other words, the intensity of a low frequency
band in a previous audio frame is highly likely not to be different
from that of a low frequency band in a current audio frame, and the
intensity of a high frequency band in the previous audio frame is
highly likely to be greatly different from that of a high frequency
band in the current audio frame.
[0038] In an audio codec where prediction coding is performed in
units of frames, only a residual value corresponding to a
difference between the intensities of a previous audio frame and a
current audio frame is encoded. Accordingly, a residual value in a
low frequency band is highly likely to be `0`, and in a high
frequency band the residual value is highly likely to be a value
other than `0`.
[0039] Similar to the intensity of the multi-channel audio signal,
the parameters thereof have different probability distributions
according to frequency bands. For example, an IID of a
multi-channel audio signal in a low frequency band is highly likely
to be `0`, and an IID of the multi-channel audio signal in a high
frequency band is highly likely to be large.
[0040] Therefore, it is inefficient to encode an intensity or a
parameter of the multi-channel audio signal that has different
probability distributions according to frequency bands by using a
single VLC table as in the related art.
[0041] For example, when Huffman coding is performed, values to be
encoded are mapped with codes having different lengths based on the
probability of occurrence of the values to be encoded. In other
words, a value less likely to be generated is mapped with a code
composed of a high number of bits, and a value highly likely to be
generated is mapped with a code composed of a small number of bits.
Accordingly, encoding of an intensity or parameter of a
multi-channel audio signal that has different probability
distributions according to frequency bands by using a single
Huffman code table may degrade the compression efficiency of the
multi-channel audio signal.
[0042] In order to address this problem, in the exemplary
embodiment of the present invention, a multi-channel audio signal
is encoded using different VLC tables according to frequency bands.
In order to achieve this encoding, the control unit 110 selects a
VLC table suitable for a frequency band that the encoding unit 120
currently encodes, and provides the selected VLC table to the
encoding unit 120.
[0043] Frequency bands are set according to sub-bands. As described
above, since encoding is performed in units of sub-bands in
parametric audio coding, frequency bands that serve as a basis for
selection of suitable VLC tables may also be set based on
sub-bands. This will now be described in greater detail with
reference to FIG. 2.
[0044] FIG. 2 illustrates frequency bands according to an
embodiment of the present invention.
[0045] As illustrated in the upper part of FIG. 2, the control unit
110 may set frequency bands serving as a basis for the selection of
VLC tables so as to be equal to a plurality of sub-bands. In other
words, when the intensities or parameters of sub-band 1, sub-band
2, sub-band 3, through to sub-band n of the multi-channel audio
signal are encoded, different VLC tables may be selected according
to the sub-bands.
[0046] In this case, a total of n.times.(p+1) VLC tables are
required, which are a sum of n VLC tables required to encode the
intensity of the multi-channel audio signal by sub-band and
n.times.p VLC tables required to encode p parameters of the
multi-channel audio signal by sub-band. In addition, if encoding is
performed using VLC tables that differ by channel, the number of
VLC tables which are referred to in order to encode a multi-channel
audio signal increases further.
[0047] In order to reduce the number of VLC tables, the control
unit 110 may define frequency bands, which serve as a basis for
selection of VLC tables, by grouping a plurality of sub-bands.
[0048] As illustrated in the lower part of FIG. 2, frequency band
1, frequency band 2, through to frequency band m may be set by
grouping sub-bands by twos. For example, a same VLC table may be
selected to encode the sub-bands 1 and 2 of the multi-channel audio
signal.
[0049] Referring back to FIG. 1, the encoding unit 120 encodes the
multi-channel audio signal in the given frequency band by using the
VLC table selected by the control unit 1 10. The intensity or
parameter of the multi-channel audio signal in the given frequency
band is encoded. Different VLC tables are selected considering
probability distributions of the intensity or parameter of the
multi-channel audio signal that differ according to frequency
bands. Thus, adaptive multi-channel audio encoding in consideration
of the different frequency bands is possible.
[0050] FIG. 3 illustrates a multi-channel audio encoding apparatus
300 according to another exemplary embodiment of the present
invention. Referring to FIG. 3, the multi-channel audio encoding
apparatus 300 includes an analog-to-digital conversion (ADC) unit
310, a control unit 320, a parameter encoding unit 330, a
down-mixing unit 340, a mono-audio encoding unit 350, and a
multiplexing unit 360. It is assumed that the multi-channel audio
encoding apparatus 300 encodes a stereo multi-channel audio
signal.
[0051] The ADC unit 310 receives an analog left-channel audio
signal and an analog right-channel audio signal of the stereo
multi-channel audio signal, samples and quantizes the analog
left-channel and right-channel audio signals, and converts the
sampled and quantized analog left-channel and right-channel audio
signals into digital signals.
[0052] The control unit 320 selects a VLC table that is to be used
to encode the multi-channel audio signal in a given frequency band.
A VLC table is selected in consideration of probability
distributions of an intensity or parameter of a mono-audio signal
in the given frequency band.
[0053] The control unit 320 has a plurality of VLC tables to be
used to encode intensities of the mono-audio signal and a plurality
of VLC tables to be used to encode parameters of the mono-audio
signal. The control unit 320 selects a VLC table that is to be used
to encode a frequency band of the multi-channel audio signal that
is being currently encoded by the parameter encoding unit 330 and
the mono-audio encoding unit 350, from two VLC tables for two
sub-bands, and provids the selected VLC table to the parameter
encoding unit 330 and the mono-audio encoding unit 350.
[0054] The parameter encoding unit 330 encodes the parameter of the
stereo-audio signal by using the VLC table selected by the control
unit 320. The parameter of the stereo-audio signal is encoded by
using VLC tables that are different according to frequency
bands.
[0055] An exemplary embodiment of the present invention provides a
method of encoding a parameter for determining the sizes of a
left-channel audio signal and a right-channel audio signal.
According to parametric audio coding, the multi-channel audio
encoding apparatus 300 down-mixes a stereo-audio signal into a
mono-audio signal and encodes the mono-audio signal, and encodes a
separate parameter for restoring the stereo-audio signal from the
mono-audio signal. In order to restore the stereo-audio signal from
the mono-audio signal, parameters for determining the phases of the
left-channel audio signal and the right-channel audio signal are
needed together with parameters for determining the intensities of
the left-channel audio signal and the right-channel audio
signal.
[0056] The IID and the IC from among the aforementioned parameter
examples are parameters for determining the intensities of the
left-channel audio signal and the right-channel audio signal.
According to the related art, two parameters for determining the
intensities of the left-channel audio signal and the right-channel
audio signal should be encoded. However, in the exemplary
embodiment of the present invention, only one parameter is used and
encoded to determine the intensities of the left-channel audio
signal and the right-channel audio signal. This will be described
in greater detail with reference to FIG. 4.
[0057] FIG. 4 is a diagram for explaining a method of encoding a
parameter for determining the intensities of a left-channel audio
signal and a right-channel audio signal, according to the exemplary
embodiment of the present invention.
[0058] A two-dimensional vector space is formed so that an L vector
associated with the intensity of the left-channel audio signal in a
given frequency band and an R vector associated with the intensity
of the right-channel audio signal in the given frequency band form
a given angle. When it is assumed that a user listens to a
stereo-audio signal at a location where directions of left and
right sound sources form a 60 degree angle, the angle between the L
and R vectors may be set to be 60 degrees in the two-dimensional
vector space. In the two-dimensional vector space formed by the L
and R vectors, an M vector associated with the intensity of the
mono-audio signal is expressed as a sum of the L and R vectors.
Since the user listens to the stereo-audio signal having an
intensity corresponding to the size of the M vector in the
direction of the M vector at the location where the directions of
the left and right sound sources form a 60 degree angle, the M
vector may be expressed as a sum of the L and R vectors.
[0059] The parameter encoding unit 330 of FIG. 3 encodes
information about an angle .theta.p between the M and L vectors or
an angle .theta.q between the M and R vectors, instead of
information about the IID and IC, as a parameter for determining
the intensities of the left-channel and right-channel audio signals
in the given frequency band.
[0060] Instead of encoding the angle .theta.p or .theta.q itself, a
cosine value such as cos .theta.p or cos .theta.q may be encoded.
When an attempt is made to encode and insert information about the
angle into a bitstream, quantization is required. Accordingly, the
cosine value of the angle is encoded in order to minimize a loss
that occurs during quantization.
[0061] The angle .theta.q between the M and R vectors has different
probability distributions according to frequency bands. This will
now be described in greater detail with reference to FIGS. 5A and
5B.
[0062] FIGS. 5A and 5B are graphs illustrating probability
distributions of an angle between vectors associated with the
intensities of a right-channel audio signal and a mono-audio
signal, that is, the angle .theta.q between the M and R
vectors.
[0063] Referring to FIG. 5A, the angle .theta.q in a low-frequency
band is highly likely to be 30 degrees. Since a stereo-audio signal
in the low-frequency band is more likely to be simultaneously
played back through two channels than to be played back through
either a left channel or a right channel, the probability that the
angle .theta.q in a low-frequency band is 30 degrees, which is half
of 60 degrees, is high.
[0064] On the other hand, referring to FIG. 5B, probabilities that
the angle .theta.q in a high-frequency band is between 0 and 60
degrees are evenly distributed. Since the stereo-audio signal in
the high-frequency band is more likely to be biased to either a
left channel or a right channel than the stereo-audio signal in the
low-frequency band, the probabilities that the angle .theta.q in
the high-frequency band is between 0 and 60 degrees are evenly
distributed.
[0065] Since the probability distribution of the angle .theta.q
varies according to frequency bands, the parameter encoding unit
330 encodes the angle .theta.q by using VLC tables different
according to frequency bands. A case where a cosine value of the
angle .theta.q instead of the angle .theta.q itself is encoded will
now be illustrated with reference to FIGS. 6A and 6B.
[0066] FIGS. 6A and 6B illustrate VLC tables according to an
embodiment of the present invention.
[0067] FIG. 6A illustrates a VLC table for encoding cosine values
of a low frequency band. Referring to FIG. 6A, in order to encode
cosine values cos .theta.q of the low-frequency band, a cosine
value cos 30.degree. is mapped to a codeword composed of the least
number of bits, and cosine values cos 0.degree. and cos 60.degree.
are mapped to a codeword composed of the most number of bits.
[0068] FIG. 6B illustrates a VLC table for encoding cosine values
of a high frequency band. Referring to FIG. 6B, a manner in which a
cosine value cos 30.degree. is mapped to a codeword composed of the
least number of bits in order to encode cosine values cos .theta.q
of the high-frequency band is the same as the manner for the
low-frequency band shown in FIG. 6A, because the probability that
the angle .theta.q is 30 degrees is highest even in the
high-frequency band as can be seen from the probability
distribution of FIG. 5B. However, the cosine value cos 30.degree.
is mapped to a 4-bit codeword, and cosine values cos 0.degree. and
cos 60.degree. are each mapped to a 6-bit codeword. In other words,
the number of bits that are used to encode the cosine values cos
0.degree. and cos 60.degree. is reduced compared with that shown in
the VLC table of FIG. 6A. Since the probabilities that the cosine
values cos 0.degree. and cos 60.degree. in the high-frequency band
are encoded are higher than in the low-frequency band, the cosine
values cos 0.degree. and cos 60.degree. in the high-frequency band
are encoded using a small numbers of bits.
[0069] The parameter for determining the intensities of the
left-channel and right-channel audio signals is only an example
taken in order to describe a method of performing adaptive encoding
on a plurality of VLC tables according to an exemplary embodiment
of the present invention. Accordingly, parameters other than the
above-exemplified parameter may also be encoded using VLC tables
that differ according to frequency bands. For example, since the
probability that a difference between phases of the left-channel
and right-channel audio signals in the low-frequency band is small
is higher than the probability that a difference between phases of
the left-channel and right-channel audio signals in the
high-frequency band is high, VLC tables that differ according to
frequency bands may be selected in consideration of this
probability distribution, and parameters may be encoded based on
the selected VLC tables.
[0070] Referring back to FIG. 3, the down-mixing unit 340 adds the
digital signals into which the ADC unit 110 has converted the
left-channel audio signal and the right-channel audio signal so as
to generate the mono-audio signal.
[0071] The mono-audio encoding unit 350 encodes the mono-audio
signal generated by the down-mixing unit 340. The mono-audio
encoding unit 350 encodes the mono-audio signal by using the VLC
tables selected by the control unit 320, which differ according to
frequency bands.
[0072] The multiplexing unit 360 multiplexes a parameter bitstream
generated by the parameter encoding unit 330 and a bitstream of the
mono-audio signal generated by the mono-audio encoding unit 350 so
as to generate a bitstream of the stereo-audio signal.
[0073] FIG. 7 is a flowchart illustrating a method of encoding a
multi-channel audio signal, according to an exemplary embodiment of
the present invention. Referring to FIG. 7, in operation 710, a
multi-channel audio encoding apparatus selects a first VLC table
that is to be used to encode the multi-channel audio signal in a
given frequency band, from among a plurality of VLC tables. Since
an intensity value or parameter value of the multi-channel audio
signal has different probability distributions according to
frequency bands, a single VLC table is selected from the VLC tables
in order to encode the multi-channel audio signal in the given
frequency band. Thus, the multi-channel audio signal is adaptively
encoded in consideration of the different probability
distributions. As described above, the VLC tables may be Huffman
code tables, and the frequency bands may be set based on
sub-bands.
[0074] Examples of a parameter of the multi-channel audio signal
may include the parameter for determining the intensities of the
left-channel audio signal and the right-channel audio signal of the
stereo-audio signal in the given frequency band. As described
above, the parameter for determining the intensities of the
left-channel and right-channel audio signals may be determined
using a vector associated with the intensity of the left-channel
audio signal and a vector associated with the intensity of the
right-channel audio signal.
[0075] In operation 720, the multi-channel audio encoding apparatus
encodes the multi-channel audio signal in the given frequency band
by using the first VLC table selected in operation 710. In other
words, in operation 720, the multi-channel audio encoding apparatus
encodes the multi-channel audio signal by using VLC tables selected
in operation 710 which differ according to frequency bands.
[0076] FIG. 8 illustrates an apparatus 800 for decoding a
multi-channel audio signal, according to an exemplary embodiment of
the present invention. Referring to FIG. 8, the apparatus 800 for
decoding the multi-channel audio signal includes a control unit 810
and a decoding unit 820.
[0077] The control unit 810 selects a variable length decode (VLD)
table that is to be used to decode the multi-channel audio signal
in a given frequency band, from among a plurality of VLD tables. In
the exemplary embodiment of the present invention, decoding is
performed using a characteristic of the multi-channel audio signal,
which varies according to frequency bands. The characteristic of
the multi-channel audio signal that varies according to frequency
bands may be a probability distribution of an intensity or
parameter value of the multi-channel audio signal. In order to
perform decoding using a characteristic of the multi-channel audio
signal, a VLD table suitable for encoding a frequency band of the
multi-channel audio signal that is being currently decoded is
selected from the VLD tables. The VLD tables may be Huffman code
tables.
[0078] The decoding unit 820 decodes the multi-channel audio signal
in the given frequency band by using the suitable VLD table
selected from among the VLD tables by the control unit 810. The
intensity or parameter of the multi-channel audio signal is
decoded.
[0079] For example, a parameter of a stereo-audio signal which is
being decoded may be a parameter for determining intensities of a
left-channel audio signal and a right-channel audio signal of the
stereo-audio signal or a parameter for determining phases of the
left-channel audio signal and the right-channel audio signal.
[0080] The parameter for determining the intensities of the
left-channel audio signal and the right-channel audio signal may be
a parameter generated by using a vector associated with the
intensity of the left-channel audio signal and a vector associated
with the intensity of the right-channel audio signal. The parameter
for determining the phases of the left-channel audio signal and the
right-channel audio signal may be a parameter associated with a
difference between the phases of the left-channel audio signal and
the right-channel audio signal.
[0081] FIG. 9 illustrates an apparatus 900 for decoding a
multi-channel audio signal, according to another exemplary
embodiment of the present invention. Referring to FIG. 9, the
apparatus 900 for decoding a multi-channel audio signal includes a
demultiplexing unit 910, a control unit 920, a parameter decoding
unit 930, a mono-audio decoding unit 940, an audio restoration unit
950, and a digital-to-analog conversion (DAC) unit 960.
[0082] The demultiplexing unit 910 receives a bitstream of the
multi-channel audio signal and separates a bitstream of a parameter
from a bitstream of a mono-audio signal.
[0083] The control unit 920 selects a VLD table that is to be used
to decode the multi-channel audio signal in a given frequency band,
from among a plurality of VLD tables. The VLD table is selected in
consideration of a probability distribution of an intensity or
parameter value of the mono-audio signal in the given frequency
band.
[0084] When the control unit 920 selects VLD tables that differ
according to frequency bands, the parameter decoding unit 930
decodes the parameter of the multi-channel audio signal by using
the selected VLD tables. Similarly, the mono-audio decoding unit
940 decodes the intensity of the mono-audio signal by using the
selected VLD tables.
[0085] The audio restoration unit 950 restores the multi-channel
audio signal based on the parameter decoded by the parameter
decoding unit 930 and the mono-audio signal decoded by the
mono-audio decoding unit 940. For example, a mono-audio signal is
transformed into a stereo-audio signal by using a parameter for
determining intensities of a left-channel audio signal and a
right-channel audio signal and a parameter for determining phases
of the left-channel audio signal and the right-channel audio
signal, the parameters having been decoded by the parameter
decoding unit 930.
[0086] The DAC unit 960 converts the stereo-audio signal restored
by the audio restoration unit 950 into an analog signal.
[0087] FIG. 10 is a flowchart of a method of decoding a
multi-channel audio signal, according to an embodiment of the
present invention.
[0088] Referring to FIG. 10, in operation 1010, a multi-channel
audio decoding apparatus according to an exemplary embodiment of
the present invention selects a VLD table that is to be used to
decode the multi-channel audio signal in a given frequency band,
from among a plurality of VLD tables. As described above, the VLD
tables may be Huffman code tables, and the frequency bands may be
set based on sub-bands.
[0089] In operation 1020, the multi-channel audio decoding
apparatus decodes the multi-channel audio signal in the given
frequency band by using the VLD table selected in operation 1010.
The intensity or parameter of the multi-channel audio signal is
decoded. Examples of a parameter of the multi-channel audio signal
may include a parameter for determining the intensity of the
multi-channel audio signal and a parameter for determining the
phase of the multi-channel audio signal.
[0090] According to an exemplary embodiment of the present
invention, a parameter of a multi-channel audio signal can be
adaptively encoded in consideration of probability distributions of
the parameter which differ according to a plurality of frequency
bands. Thus, the multi-channel audio signal can be encoded at a
compression rate higher than that in a related art.
[0091] Moreover, information about intensities of a left-channel
audio signal and a right-channel audio signal of a stereo-audio
signal can be encoded using a smaller number of parameters than in
the related art. Thus, the multi-channel audio signal can be
encoded at a compression rate higher than in the related art.
[0092] The invention can also be embodied as computer readable
codes on a computer readable recording medium. The computer
readable recording medium is any data storage device that can store
data which can be thereafter read by a computer system. Examples of
the computer readable recording medium include read-only memory
(ROM), random-access memory (RAM), CD-ROMs, magnetic tapes, floppy
disks, optical data storage devices, and carrier waves (such as
data transmission through the Internet). The computer readable
recording medium can also be distributed over network coupled
computer systems so that the computer readable code is stored and
executed in a distributed fashion.
[0093] While the present invention has been particularly shown and
described with reference to exemplary embodiments thereof, it will
be understood by those of ordinary skill in the art that various
changes in form and details may be made therein without departing
from the spirit and scope of the present invention as defined by
the following claims.
* * * * *